IR 05000387/1989030
| ML17156B527 | |
| Person / Time | |
|---|---|
| Site: | Susquehanna  |
| Issue date: | 11/24/1989 |
| From: | Swetland P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML17156B526 | List: |
| References | |
| 50-387-89-30, 50-388-89-27, NUDOCS 8912120107 | |
| Download: ML17156B527 (21) | |
Text
U.
S.
NUCLEAR REGULATORY COMMISSION
REGION I
Report Nos.
50-387/89-30 50-388/89-27 License Nos.
NPF-14 NPF-22 Licensee:
Penns lvania Power and Li ht Com an 2 North Ninth Street Allentown Penn s 1 vani a 18101 Facility Name:
Sus uehanna Steam Electric Station Inspection At: Salem Townshi Penns lvania Date:
Se tember
1989 October
1989 Inspectors:
G ~
S. Barber, Senior Resident Inspector, SSES J.
R. Stair, Resident Inspector, SSES C.
H.
oodard, React Engineer, DRS Approved By P.
Swetland, Chief Reactor Projects Section No.
2A, D te Ins ection Summar Areas Ins ected:
Routine inspections were conducted in the following areas:
plant operations, physical security, plant events, surveillance, and maintenance.
Results:
During this period, Operations Department personnel generally conducted activities in a professional manner and operated the plant safely.
Routine review of maintenance and. surveillance activities noted good control and performance.
A spill of approximately 50 gallons of radioactive water occurred in the Unit 2 drywell as a result of an inadequately secured drainline during LPRM change-out.
Licensee corrective actions were timely and comprehensive.
Two drywell purge isolation valve closures occurred as a result of failed solenoids. 'he required 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> notification for the first isolation was made 3'ours l,ate'.
This is a licensee-identified violation for which no Notice of Violation wi.ll be issued, but corrective action is needed to prevent recurrence.
8~>gggOi07 ~50'()~87 PDR ADOCN 0 pt4U
I
'n unplanned full scram signal occurred in Unit 2 due to a failed air line connection at a Swagelok fitting.
No rod motion occurred since the plant was
'hut down with all rods inserted.
The licensee's corrective actions in response to the event were appropriate.
General Employee and Health Physics II annual retraining classes were audited by the inspector.
The retraining was considered to be generally effective.
Two Emergency Diesel Generator (EDG) Crankcase ignitions and overpressuriza-tions occurred during the performance of 18 month load rejection surveillance tests.
Both failures occurred at a test imposed overload condition of 4700 kilowatts.
Both EDGs were rep'aired and returned to operable status following successful completion of the 18 month surveillance tests.
A root cause determination is pending metallurgical analyses of the damaged pistons and liner TABLE OF CONTENTS l
Page 1.0 Introduction and Overview
.
.
.
.
.
.
.
.
.
.
~
.
.
.
.
.
.
.
1.1 NRC Staff Activities (30703, 71707, 90712, 92701, 60710).
1.2 Unit 1 Summary.
.
.
.
,
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1.3 Unit 2 Summary.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2.0 Routine Periodic Inspections (93702)
~
~
~
2.1 Scope of Review
.
2.2 Drywell Spill - Temporary Work Stoppage Unit 2.
2.3 Unplanned Reactor Scram While Shutdown - Unit 2
.
2.4 Automatic Isolation Valve Closures - Unit 2
.
2.5 Division I - 24 Volt Battery Inoperability - Unit 2
~
~
~
~
~
3
~
~
~
3.0 Surveillance and Maintenance Activities.
.
.'
.
.
.
.
.
.
.
.
". 1 Surveillance Observations (61726)
3.2 Maintenance Observations (62703)
~
'I 4.0 Licensee Reports
.
7
4. 1 Review of Licensee Event Reports (92700).
.
.
.
~
.
.
.
.
4.2 Review of Significant Operating Occurrence Reports (90712)
5.0 Access Training (83523)
6.0 Emergency Diesel Generator Failures (93702 and 71707).
.
.
.
.
6. 1
"B" Emergency Diesel Crankcase Overpressurization 6.2
"C" Emergency Diesel Crankcase Overpressurization
13 7.0 Resident Monthly Exit Meeting (30703)
.
.
.
.
.
.
.
.
.
.
.
.
DETAILS Introduction and Overview NRC Staff Activities The purpose of this inspection was to assess licensee activities at the Susquehanna Steam Electric Station (SSES)
as they related to reactor safety and worker radiation protection.
Within each area, the inspectors documented the specific purpose of the area under review and the scope of inspection activities and findings, along with appropriate conclu-sions.
This assessment is based on actual observation of licensee activities, interviews with licensee personnel, measurement of radiation levels and independent calculations, when appropriate, and selective review of applicable documents.
Unit 1 Summar Unit 1 entered the inspection period in cold shutdown for the installa-tion of orifices in the suppression pool to drywell vacuum breaker relief valve test circuits.
The details were discussed in Inspection Report 50-387/89-24.
The turbine was paralleled to the grid on September 12.
Difficulty in opening the "B" reactor feedpump discharge valve limited power to 75 percent until September 17 when the valve was opened and full power operation resumed.
The unit then operated at full power for the remainder of the inspection period.
Two Diesel Generator crankcase explosions occurred during 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> required surveillance runs on the "B" and "C" Diesel Generators.
See Section 6.0 for details.
Unit 2 Summar Unit 2 remained in Condition 5 for its third refueling outage which began on September 10.
During the inspection period major activities included total core offload which was completed on September 27, replacement of 24 control'od drive mechanisms and 20 Local Power Range Monitor (LPRM) Strings, performance of reactor pressure vessel internal exams and, completion of several special/major projects.
Core reload commenced on October 12 and included 204 new fuel bundles.
On September 27, a spill in the drywell occurred during LPRM removal.
See Section 2.2 for details.
On October 8, an unplanned scram signal occurred due to a broken air line to a scram pilot solenoid valve.
See Section 2.3 for details.
Drywell Purge Isolation Valve closures occurred on September 10 and 11.
See Section 2.4 for details.
Overall, outage activities have proceeded smoothly and on schedul.0 Routine Periodic Ins ections 2. 1 Sco e of Review The facility was periodically inspected to determine the licensee's compliance with the general operating requirements of the Technical Specifications (TS) in the following areas:
review of selected plant parameters for abnormal trends; plant status from a maintenance/modification viewpoint, including plant housekeeping and fire protection measures; control of ongoing and special evolutions, including control room personnel awareness of these evolutions; control of documents, including logkeeping practices; implementation of radiological controls; implementation of the security plan, including access control, barrier integrity, and badging practices; control room operations during regular and backshift hours, including frequent observation of activities in progress, and periodic reviews of selected sections of the unit supervisor's log, the control room operator's log and other control room daily logs; followup items on activities that could affect plant safety or impact plant operations; areas outside the control room; and, I
selected licensee planning meetings.
The inspectors conducted backshift and weekend/holiday inspections on October 3, from 2:30 a.m. to 6:00 a.m.,
and October 11, from 2:30 a.m.
to 6:00 a.m..
The inspectors reviewed the following specific items in more detail.
2.2 Dr well S ill - Tem orar Work Sto a
e - Unit 2 On September 27, 1989 the licensee identified that a 50 gallon spill occurred in the drywell at 8:00 p.m. during Local Power Range Monitor (LPRM) changeout.
An LPRM was being backflushed through a temporary drain line (temporary hose) that was lined up to a drywell downcomer.
The hose worked its way out of the downcomer and sprayed the 704 foot elevation of the drywell.
Approximately 50 gallons of water sprayed a
12 foot by 30 foot area and the licensee stopped all work in the drywell
until the area could be decontaminated.
No personnel contaminations resulted from the spill.
Airborne contamination generated from the spill was minimized and equated to 18 percent MPC in the local area during decontamination.
The licensee decontaminated the area and resumed work on September 28.
The licensee determined that the event was not reportable.
Discussion with the licensee determined that inadequate procedural guidance led to the hose being improperly inserted into the downcomer unsecured.
The'emporary hose is normally installed with health physics support but in this case, Instrument and Control (ICC) technicians performed the evolution without health physics guidance.
The hose worked its way out of the drain line and sprayed the area.
The event could have been prevented if the IKC technicians had more experience with the evolution or if the -procedure required securing the hose end.
The procedure which controls the evolution RE-081-030, LPRM Installation-and Removal, has since been modified to ensure that the hose (Tygon tubing) from an LPRN seal tube is securely attached to the suppression pool downcomer and inserted at least six feet.
The inspectors reviewed the event for reportability and determined that although the licensee temporarily stopped work in order to effect cleanup, the event did not meet the criteria necessary fo'r reporting under
CFR 50.72, 50.73, or 20.403.
The inspector considered the licensee's actions in response to the event appropriate and conserva-tive.
2.3 Un lanned Reactor Scram While Shutdown Unit 2 The licensee reported that an unplanned full scram signal was generated at 7:41 a.m.,
October 8.
The signal was generated when a broken air line to.the scram pilot solenoid valve for control rod drive (CRO) 26-47 depressurized the air header causing the scram valves for three other CROs to open and fill the Scram Discharge Volume (SDV).
When the SDV reached its high level setpoint, a full scram signal was generated.
No rod motion occurred since all CRDs were fully inserted.
The scram pilot valve for CRO 26-47 was isolated to repressurize the air header and to reset the scram signal.
The scram signal was reset.
The event was reported to the NRC at 11:00 a.m., October 8 in accordance with 10 CFR 50.72(b)(2)(ii).
The licensee concluded that the root cause of this event was the improper make-up of a tubing-to-Swagelok fitting in the instrument air supply to the scram pilot solenoid valve for CRO 26-47.
The fitting had not been purposely manipulated during any of the routine changeouts of the scram pilot solenoid valves.
The licensee surmised that since the fitting is in close proximity to the solenoid valve for CRD 26-47 (about 6 inches),
a change-out of the solenoid valve on July 26, 1989 could have contributed to the fai lure due to stressing the tubing-to-fitting connectio The licensee plans to modify the Instrument and Control post maintenance test procedure, IC-055-002 (Hydraulic Control Unit Scram Pilot Solenoid Valve Rework), to leak check all fittings associated with each scram pilot solenoid valve after the air supply is restored to the Hydraulic Control Units following maintenance.
Also, maintenance personnel performing this work will receive appropriate training to assure that they are familiar with the leak check procedure.
The, inspector will review the LER when it is. issued.
2.4 Automatic Isolation Valve Closures - Unit 2 During a drywell purge to de-inert the primary containment, on September 10 at 3:00'.m., drywell air purge outboard isolation valve (HV-25723)
open indication deenergized and the valve appeared to close.
Investiga-tion by the licensee determined that the valve solenoid operator had failed, causing its control power fuse to blow.
The blown fuse resulted in the loss.of valve position indication and valve closure.
The licensee entered the Technical Specification limiting condition for operation and isolated the affected penetration until repairs were made following replacement of the blown fuse and the failed solenoid.
The drywell purge was reestablished at 4:00 a.m.
on September ll.
llowever, at 6:00 a.m., during a panel walkdown operators-observed that HV-25723 again indicated full closed.
Attempts to reopen the valve were unsuccess-.
ful.
The remaining valves in the purge alignment were again closed and an investigation performed to determine the cause.
The licensee determined that the solenoid which had been r'ecently replaced also failed.
Since the unit had entered a refueling, outage at the time, the valve was kept isolated until it could be reworked and demonstrated operable prior to startup from the refueling outage.
The licensee determined that the cause of both the valve closures was due to failed solenoids which were replaced.
Additionally, the licensee is evaluating the event to determine whether these failures are isolated cases or are indicative of a more generic problem.
The valve has since been returned to service with no recurrent failures NRC will review the generic aspects of this event upon receipt of the Licensee Event Report.
The licensee determined that an Emergency Notification System (ENS) call
~to the NRC was required due to the automatic isolation of this contain-ment isolation valve.
Both closures,resulted in ENS calls, however, the first call was not made within the required 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> time, but was made almost 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> late.
This was identified by the licensee and was brought to the attention of the NRC.
The failure to report the first isolation within the required 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> time period is considered a
licensee identified violation per
CFR 2.
The Licensee Event Report (LER) stated that licensed operators are being briefed on the subjects of ESF actuation recognition and specific actions required to be taken when one is determined or suspected to have occurred.
In addition, enh'ancements concerning this subject are to be made to the formal operator training programs.
(NV5 50-388/89-27-02).
2,5 Division I - 24 Volt Batter Ino erabilit Unit 2 Two 24-volt batteries provide. power to Division (Div.) I DC loads..
These two batteries 'are electrically configured to appear like one
volt battery (2D670) that is center.tapped to form a plus (+) and a
minus (-) 24 volt battery.
These two batteries are supplied by two separate battery chargers which normally supply DC loads since their output voltage is greater than that.of the batteries.
The Process Radiation Monitors, Source Range Monitors (SRMs)
and Intermediate Range Monitors (IRMs) are supplied by the
VDC system.
The licensee declared the 2D670 battery bank and the components it powers inoperable at 11:30 a.m.,
October 13.
Twelve cells of 2D670 registered a specific gravity of 1. 170 which was less 'than the specific gravity limit of 1. 195 imposed by Technical Specifications (TSs).
The licensee ceased fuel movement due to the SRMs being declared inoperable in,.accordance with TS action statement-3'.2.
The licensee informed the inspector of the inoperabi)ity on October 13.
The licensee also informed the inspector that the SRMs were functioning normally because the battery charger was supplyi,ig them along with charging the battery.
The licensee clarified that only the "+" 24 volt battery had low specific gravity.
Fuses were pulled to separate the battery halves.
The inspector noted that the SRMs that were being powered by the inoperable (+24 volt) battery half continued to function normally even after 'electrical separation.
The licensee believed that the loss of the battery was insignificant because the SRMs that powered it 'continued to function.
The licensee believed that the
VDC system for these SRMs should be considered operable.
The licensee revi'ewed the definition of "operabi.lity" along with a clarification of the "operability" definition provided in an August 21, 1989 letter from the Director, Nuclear Reactor Regulation (NRR) to the Senior Vice President (VP) of Illinois Power Company (Clinton Power Station/50-461).
The lett'er stated the following:
"The definition of operability affects the manner in which the require-ments for a Limiting Condition for Operation (LCO) and its associated remedial actions are applied when a support system is inoperable.
If the licensee determines that a
TS system is capable of performing its intended function with an inoperable support system, then no additional action is needed.
If the licensee determines that a
TS system could not perform its intended safety function with an inoperble support system, then the TS LCO must be entered and appropriate remedial actions taken.
This action shall'occur regardless of whether or not the support system is covered by TS."
Since this was a previously accepted staff position on operability, the licensee determined it was applicable to their situation regarding the Div. I-24 volt battery banks and chargers.
Since the SRMs were powered directly by the operable battery charger without the need for the support action of the battery, they considered them operable.
The lic'ensee's analysis was placed in their Unit 2 Technical Specifications
.
(TS) Interpretations Manual.
The inspector noted that the licensee electrically separated the "+" and
"-" 24 volt batteries to minimize the effects of, the undercharged cells.
Also, the inspector noted the "operability" clarification as a previously accepted staff position.
In addition, the inspector noted that the licensee would be able to.provide immediate remedial actions by ceasing core alterations if a future 24 VDC problem occurred.
Therefore, the inspector noted no inadequacies with the licensee's operability deter-minations~
The "+" 24 volt battery was subsequently fully recharged and returned to an operable status.
The inspector had no further questions on this issue.
3.0 Surveillance and Maintenance Activities On a sampling basis, the inspector observed and/or reviewed selected surveillance and maintenance activities to ensure that specific programmatic elements described below were being met.
Details of this review are documented in the following sections'.1 Surveillance Observations The inspector observed and/or reviewed the following surveillance tests to determine that these criteria, if applicable to the specific test, were met:
the test conformed to Technical Specification requirements; administrative approvals and tagouts were obtained before initiating the surveillance; testing was accomplished by qualified personnel in accor-dance with an approved 'procedure; test instrumentation was calibrated; Limiting Conditions for Operations were met; test data was accurate and complete; removal and restoration of the affected components was pro-perly accomplished; test results met Technical Specification and proce-dural requirements; deficiencies noted were reviewed and appropriately resolved; and the surveillance was completed at the required frequency.
These observations and/or reviews included:
SE-253-301, Standby Liquid Control (SLC) Functional Leak Test, performed on October 5, 1989.
TP-225-015, Pressure Test of the Containment Instrument-Gas 2200 psig/150 psig Header in support of the Freon Flush, performed on October 5,
198 S0-151-002, quarterly Core Spray Flow Verification, performed on October 13, 1989.
S0-251-014, Core Spray System Cold Shutdown Valve Exercising, performed on October 13, 1989.
No unacceptable conditions were identified.
3.2 Maintenance Observations The inspector observed and/or reviewed selected maintenance activities to determine that the work was conducted in accordance with approved procedures, regulatory guides, Technical Specifications, and industry codes or standards.
The following items were considered, as applicable, during this review:
Limiting Conditions for Operation were met while components or systems were removed from service; required administrative approvals were obtained prior to initiating the work; activities were accomplished using approved procedures and quality control hold points were established where required; functional testing was performed prior to declaring the involved component(s)
opera'e; activities were accomplished by qualified personnel; radiological controls were implemented; fire protection controls were implemented; and the equipment was verified to be proper ly returned to service.
These observation's and/or reviews included:
Replacement of the "B" Diesel Generator 7L and 7R pistons and cylinder liners per Work Authorization (WA) S94568, on September 18, 1989.
Installation of panel insert, switches, isolation cans, and indicating lights in panel 2C601-18C per Construction Work Order (CWO) C90507, on October 5, 1989.'eworking of panel wiring and termination of field cables EK2Y0030 B/D in panel 2C601 per CWO C99557, on October 5, 1989.
Investigation of the cause of the crankcase vapor ignition on the
"C" Diesel Generator per WA S94647, on October 10, 1989.
Removal of the head and replacement of the 5R piston and liner on the "C" Diesel Generator per WA S94649, on October 10, 1989.
No inadequacies were note S 4.0 Licensee Re ort's 4.1, Review of Licensee Event Re 'orts The inspector reviewed LERs submitted to the NRC office to verify that details of the event were clearly reported, including the accuracy of the description of the cause and the adequacy of corrective action.
The inspector determined whether further information was required from the licensee, whether generic implications were involved, and whether the event warranted onsite followup. The following LERs were reviewed:
Unit
89-023-00 Plant Shutdown completed when Vacuum Relief Valves Declared Inoperable -- Event Detailed in NRC Inspection Report 50-387/89-24.
Unit 2 89-006-00 Unplanned Exposure Potentially in Excess of 10CFR20 limits - Event Detailed in NRC Inspection Report 50-388/89-20
'9-007-00 Primary Containment Isolation Valve Automatic Closure-
'Event-Detailed in NRC Inspection Report 50-388/89-22.
No inadequacies were noted.
. 4.2 Review of Si nificant 0 eratin Occurrence Re orts Significant Operating Occurrence Reports (SOORs) are provided for problem identification tracking, short and long term corrective actions, and reportability evaluations.
The licensee uses SOORs to document and bring to closure problems identified that do not merit an LER.
The inspectors reviewed the following SOORs during the period to ascertain whether:
additional reactive inspection effort or other NRC response is warranted; corrective action discussed in the licensee's report appears appropriate; generic issues are assessed; and, prompt notification was made if required:
1'-89-304, 1-89-306, 1-89-308, 1-89-310, 1-89-311, 1"89-312, 1-89-313, 1-89-314, 1-89-315, 1-89-316, 1-89-317, 1-89-318, 1-89-319, 1-89-320,
'2-89-111, 2-89-113, 2-89-116, 2-89-117, 2-89-118, 2-89-119, 2-89-120, 2-89-122, 2-89-123, 2-89,124, 2-89-125, 2-89-126, 2-89-127, 2-89-128, 2-89-129, 2-89-130, 2-89-131, 2-89-132, 2-89"133, 2-89-134, 2-89-135, 2-89-136, 2-89"137, 2-89-139, 2-89-140, 2-89"141, 2-89-142, 2-89"143, 2-89-144, 2-89-145, 2-89-146, 2-89-148, 2-89-149.,
and, 2-89-155.
The following SOORs required inspector followup:
1-89-309, documented a crankcase vapor ignition in the "B" diesel generator.
See Section 6.0 for details.
2-89-138, documented a
50 gallon spill in the drywell during LPRM changeout.
See Section 2.2 for details.
2-89-153, documented an Unplanned Automatic Scram Signal.
See Section 2.3 for details.
2-89-115, documented the automatic closure of the drywell purge outboard isolation valve.
See Section 2.4 for details.
2-89-161 documented the inoperability of the Division I 24 VDC Batteries due to specific gravity below the Technical Specifications limits.
See Section 2.5 for details'.0 Access Trainin On September 21, 1989, the inspector attended General Employee (GET)
and Health Physics II (HP2) annual retraining in order to audit this area of the licensee's training program.
The retraining was intended to refresh individuals knowledge level in areas, such as, controlled zones, vital and protected area controls, badging, access and egress control, plant alarms, permit and tag sy'tem, protective equipment, the right to know law, fire protection, paths available for resolving safety concerns, ALARA, fitness for duty, Radiation Work Permits, and, additional health physics aspects.
The inspector audited the course to review the adequacy of training in these areas.
The inspector noted that the students performed a stay time calculation using a health physics survey map and an RWP during the class.
The exams were fairly comprehensive and covered the objectives presented.
The exam questions and answers appeared to be clear and easy to understand.
The inspector also noted that the instructor covered the objectives and the lesson plans very quickly.
The inspector was concerned that the rapid pace of the class may have been too quick for the slower students.
The inspector noted that all the individuals in the class passed the GET retraining exam, but three out of 15 students failed the HP2 retraining exam.
Discussion with the instructor following the class sessions indicated that the retraining philosophy intends that an individual who cannot pass the retraining be required to attend the initial GET and HP2 training sessions.
This assures that only individuals capable of successfully completing the exams remain qualified.
The inspector considered the retraining to be generally effective'
6.0 Emer enc Diesel Generator Failures 6. 1
"B" Emer enc Diesel Crankcase Over ressurization The licensee reported that the "B" Emergency Diesel Generator (DG)
crankcase ignited and overpressurized during a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> run at 9:50 a.m.,
September 16.
An operator had just finished making rounds on the DG skid and was on his way to check the DG's ventilation lineup when he heard a loud popping noi se and the room filled with smoke.
He quickly exited the area and called the control room to shutdown the DG.
Control room operators used normal shutdown procedures to secure the diesel since no emergency shutdown capability existed in the control room.
The diesel continued to run for an additional 4 minutes as a part of its normal cool down sequence.
After the DG shutdown, a fire brigade entry was made to investigate the damage.
An operator wearing a Scott Air Pack entered the smoke filled compartment and noted that a "crankcase high pressure" alarm was in on the local control panel.
After his exit, the area was successfully ventilated.
Further investigation rev'ealed that the No.
7L cylinder liner and piston were severely damaged.
The resultant metal-to-metal contact generated enough heat and sparking to ignite vapors in the crankcase.
Licensee examination following the shutdown provided the following information:
A crankcase vapor ignition occurred in which the crankcase explosion relief door s opened.
Oil vapors and smoke filled the DG bay.
No fires or secondary explosions occurred.
Visual inspection showed that the number 7 left piston and cylinder liner had been extensively damaged and were considered the source of the excessive heat which led to the crankcase ignition.
The 7L piston pin and connecting rod bolts were tight (properly torqued).
Removal/disassembly of the 7L piston, piston pin, and cylinder liner disclosed the following:
There was extensive evidence of piston overheating and scuffing within the cylinder liners, and there was evidence of severe piston overheating and metal deformation on the thrust side of the piston skirt in the vicinity of the bottom oil ring.
There was also severe scuffing and overheating of one of the piston pin end caps due to rubbing the cylinder liner.
A crack existed in the piston skirt beneath the damaged piston pin end cap progressing from the skirt bottom edge'pward approximately two inches.
The cylinder liner was badly scored with considerable piston metal deposits at the rubbing interfaces.
The cylinder liner jacket water expansion joint was severely deformed due to the flashing of retained water under high heat load The number 7L connecting rod and associated bearings were not damaged.
The number 7R (opposite side) cylinder liner had minor piston pin end cap rubbing indications.
The number 7L piston bronze bearing insert was not frozen to the piston pin.
After separation, the bearing showed some discoloration evidence of overheating on the thrust side.
The piston pin showed little evidence of any damage.
Engine parameters including temperatures recorded ten minutes before the explosion'ere uniform and normal.
The "B" DG had accumulated a total of 1033. 1 running hours with 11.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> of operation since the, last 18 month inspection.
Spectrographic sample analyses of engine oil taken during recent runs did not disclose any unusual contamination with metal particulates.
The licensee, with the assistance of DG vendor personnel, is conducting a root cause analysis in order to pinpoint the root cause of the engine failure.
The damaged piston is being sectioned for a detailed metal-lurgical analysis.
During this inspection, the following observations were made:
The "B" DG is a
16 cylinder Cooper-Bessemer KSV-16 Series diesel generator,
.as are the "A", "C", "D" DGs.
The "E" DG is a 20 cylinder Cooper KSV-20 Series diesel generator.
The diesels of interest, the KSV-16s, are rated at 5580HP with a peak horsepower (HP) rating of 6556HP.
The generator, itself, is rated at 4000 KM.
The KSV-16 engine is a V-16 cylinder design with a main and an articulated connecting rod for each crankshaft journal such that a right and left side connecting rod both provide power to the same crankshaft journal.
In each of the three crankcase ignitions at Susquehanna, the source of excessive heat has always been from a left side cylinder (articulating rod side).
In this engine, the piston pins (wrist pins) are bolted to the connecting rods and therefore are an integral part of the connecting rod and not the piston.
The pistons have piston pin end covers pressed into the piston pin end holes.
These covers are machined with the piston curvature to form a continuous piston surface except for the press-fit junctio Two previous,DG crankcase ignitions have occurred at Susquehanna.
During 1984, the "D" DG 2L piston and liner were involved.
During 1986, an ignition involved the "B" DG 5L piston and liner.
One of the events was attributed to loss of lubrication due to loose piston pin bolts and the other to cylinder scuffing by the piston pin end cap and rings.
For the current event concerning the "B" DG, the crankcase explosion relief covers appeared to have functioned properly to relieve the overpressure and to reseat quickly in such a manner that sufficient air was not drawn into the crankcase to cause a
secondary explosion and/or fire.
The purpose of the piston pin end cap covers was reviewed by the inspector.
They appear to serve no functional purpose, as a part of the piston, and have an established history of moving out of their pressed-in location to engage the cylinder liner and rub.
According to the on-site vendor technical representative, it is not unusual for these caps to "walk-out" and rub the cylinder liner.
Examination of the No.
7L piston bronze insert bearing wear pattern for the piston pin did not provide evidence of piston longitudinal movement sufficient to force the end caps against the cylinder.
The licensee committed to making an end cap evaluation with vendor assistance to determine the potential causes'of the failues, industry experience with end cap pr'oblems and to make a determination of what changes, if any, will be made in the equipment or procedures by November 30, 1989.
Since the licensee conducts spectrographic analysis of oil samples for metals contamination for predictive indication of pending engine problems, the sampling location and method of sampling are important.
To be meaningful the oil samples should be taken from a running engine from a sampling location which provide samples most representative of oil in the crankcase.
This inspection revealed that the oil samples are taken from a shutdown DG at a location downstream of the oil strainers and oil filters (which would filter out metal particulates)
just prior to the oil returning to the crankcase.
The licensee intends to change the oil sampling procedure and oil sampling location to ensure a representative sample was taken near the end of its run and to take the sample from an oil piping location as it comes out of the engine crankcase ahead of any filtration.
Further, the licensee will begin trending of the sample analysis as an aid in predictive maintenance'icensee implementation dates for these changes were not establishe The danger of fire from a crankcase ignition and the possibility of additional engine damage is considerably greater if the engine continues to run following the ignition.
The licensee's design does not provide a
means outside the DG room for quickly shutting down a
DG without the cooldown cycle.
Because of the personnel safety, fire hazard and equipment damage implications, the licensee concurred that a modifica-tion was needed to provide this shutdown capability.
An additional concern was expressed to the licensee regarding the 125VDC control circuit for the DG units.
The design is such that upon loss of, 125VDC to this circuit an operating DG will shutdown.
Such a shutdown during an accident scenario which requires the DG contributes to the DG unreliability.
However, if tPe loss of 125VDC causes the operating DG to shutdown when it is synchronized with the grid (monthly test),
the grid will power the generator as a motor which attempts to drive the DG.
The reverse power will not trip the DG circuit breaker because it depends upon the same 125VDC circuit which was lost.
An incident like this occurred at another Region I plant with similar DG units.
During that event the generator w'as extensively damaged due to overheating and fire.
The licensee indicated that a study of the 125VDC circuitry for the DG and associated equipments will be made to determine
= course of action.
6.2
"C" Emer enc Diesel Crankcase Over ressurization The licensee reported that the "C" Diesel Generator (DG) crankcase ignited and overpressurized during its 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> surveillance run at 11:03 p.m., October 7.
The diesel was started at 10:28 p.m.
and ran for 35 minutes at 4700KW when a diesel low priority alarm and a fire alarm were received in the control room.
A Nuclear Plant Operator (NPO) was sent to investigate and reported smoke coming from under the "C" DG enclosure doorway.
Control room operators used normal shutdown procedures to secure the diesel since,no emergency shutdown capability exists in the control room.
The diesel continued to'run for approximately 4 to 5 more minutes as a part of its normal cool down sequence.
After the DG shut-
.
down, a fire brigade entry was made to investigate the damage.
An operator wearing a 'Scott Air Pack entered the area and noted a "crank-case high pressure" alarm on the local control panel.
After.his exit, the area was successfully ventilated.
Further investigation revealed that the No.
SR cylinder piston pin had seized.
This caused the piston to side load the cylinder wall.
This, in turn, caused it to crab up and down in the cylinder which caused direct metal-to-metal contact between the piston's tin coating and the chrome cylinder wall.
The resultant metal-to-metal contact generated enough heat and sparking to ignite vapors in the crankcase.
Licensee examination following the shutdown provided the following information:
The number 5R piston and cylinder liner had received extensive damage and were considered the source of the excessive heat which led to the explosio The 5R piston pin, connecting rod, and wrist pin bolts were tight (properly torqu'ed).
The 6L piston and cylinder liner showed evidence of piston slap, tin deposits, and scuffing on the inboard side.
Removal/disassembly of the 5R piston and cylinder liner disclosed the following:
Extensive evidence of piston pin overheating.
The hardened steel piston pin was badly discolored (blued) from overheating.
It appeared to have seized within the piston bronze bushing which caused the bushing to, rotate approximately 5 degrees out of place.
Evidence of piston overheating and metal deformation on both the thrust and inboard sides of the piston.
A large through wall crack in the piston skirt beneath an end cap (approximately'he same location as the crack in the "B" EDG piston).
The crack extended from the bottom of the'kirt diagonally upward about 3 inches.
Two piston skirt casting surface defects (cavities) about I/4-inch diameter by 3/16-inch deep in the lower skirt about 3/4-inch from the lower edge of the skirt and about 2-inches apart.
There was a hair line crack from one of the cavities to the lower edge of the piston skirt.
This crack was about 2-inches from the large crack.
The piston pin boss beneath the piston was cracked from the piston pin hole into the boss approximately one-inch.
The bound piston pin and bronze bushing were removed from the cylinder by sawing a small gap longitudinally from one end to the other's The pin then easily came free of the bronze bearing.
A substantial amount of lube oil (from the pre-lube system)
was found within the pin/sleeve which indicated that the lubrication passages were still open following the explosion and engine shutdown.
The bronze bushing/bearing had a crack from one end through to the connecting rod opening in the bearing.
The pistons of the DG are tin-plated.
According to the vendor, the tin plating provides a bearing surface for the cast-iron piston in the steel cylinder liner.
During initial operation of the OG, some piston slap with associated tin rub off onto the liner is not unusual.
Following break-in continued engine slap with additional piston slap/scuffing is not expected.
The vendor recently provided the licensee with acceptance/rejection criteria for the amount of piston slap/tin deposits for the cylinders.
As observed the number 6L piston cylinder liner had an area of
r
<
.
approximately 5 inches wide by 8 inches long of tin deposit/scuffing from piston slap.
This was considered to be acceptable.
However, the cyl inder 1 incr'tin plate out was removed with scotch brite abrasive and the liner was re-installed with a new piston.
Bearing loading is directly related to the peak firing pressure in each cylinder.
The engine manual states that the standard fuel injection timing is 28 degrees before top dead center.
It'urther states that the fuel injectors timing can be retarded if necessary to reduce firing pressure.
The vendor field service representative and the licensee stated that the fuel injector timing had previously.been retarded on at least one of the engines to reduce (balance)
cylinder firing pressures.
The licensee does not record engine cylinder firing pressures.
However, individual cylinder exhaust temperatures are recorded during surveillance runs which can be related to the cylinder firing pressure.
These temperatures appear to be uniform and normal in the readings taken before these events.
The function'f the oil scraper ring on the bottom of the. piston skirt is not well understood.
It appears that it could reduce piston/cylinder liner lubrication by preventing the oil rich crankcase splash and vapors from reaching a substantial portion of the cylinder liner area.
Further, tin deposits from piston slap are scrubbed by this bottom oil scraper ring and could contribute to ring fowling, heating and additional bearing loading.
In both the "B" and "C" ignition events, areas of the piston/cylinder liners in the vicinity of the piston bottom scraper ring were extensively damaged including piston metal flow over the piston ring.
-'In order to provide for proper cylinder compression with minimum exhaust blow by, piston compression ring gaps are normally staggered around the piston.
The inspector observed that the gaps of the four compression rings were essentially in-line which would permit maximum exhaust gas blow-by into the crankcase.
The "C" DG had accumulated a total of 796.7 running hours with 26.7 operating hours since the last 18 month inspection.
The inspector reviewed the event in detail with the licensee.
Conference calls were held between the NRC and the licensee to provide reasonable assurance that the immediate failure mechanisms were not inherent to the design of the Cooper-Bessemer KSV-16 series diesel generator.
Vendor representatives investigated the failures onsite and were present for consultation during the repair efforts.
The licensee is performing a detailed set of metallurgical analyses to determine the root cause.
The vendor believes both failures originated in the piston skirt area.
The licensee is considering this cause along with the potential for inadequate individual cylinder lubrication under high
loads (-4700 KW or 118 percent of rated load).
The licensee expects to complete the metallurgical exams for.,'both failures and issue the final report by February 1990; The root cause determination and adequacy of long term corrective actions for these diesel crankcase ignitions is unresolved pending NRC review of the licensee's submittals.
(UNR 50-387/89-30-01)
(Common).
7.0 Resident Monthl Exit Meetin At the conclusion of the inspection period, the inspector discussed the findings of this inspection with station management.
Based on NRC Region I review. of this report-and discussions held with licensee representatives, it was determined that this report does not contain information subject to
CFR 2.790 restriction p+ c
'
I I gj'